Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387639503> ?p ?o ?g. }
- W4387639503 abstract "Abstract Autoregressive Neural Networks (ARNNs) have shown exceptional results in generation tasks across image, language, and scientific domains. Despite their success, ARNN architectures often operate as black boxes without a clear connection to underlying physics or statistical models. This research derives an exact mapping of the Boltzmann distribution of binary pairwise interacting systems in autoregressive form. The parameters of the ARNN are directly related to the Hamiltonian’s couplings and external fields, and commonly used structures like residual connections and recurrent architecture emerge from the derivation. This explicit formulation leverages statistical physics techniques to derive ARNNs for specific systems. Using the Curie–Weiss and Sherrington–Kirkpatrick models as examples, the proposed architectures show superior performance in replicating the associated Boltzmann distributions compared to commonly used designs. The findings foster a deeper connection between physical systems and neural network design, paving the way for tailored architectures and providing a physical lens to interpret existing ones." @default.
- W4387639503 created "2023-10-15" @default.
- W4387639503 creator A5046876703 @default.
- W4387639503 date "2023-10-14" @default.
- W4387639503 modified "2023-10-15" @default.
- W4387639503 title "The autoregressive neural network architecture of the Boltzmann distribution of pairwise interacting spins systems" @default.
- W4387639503 cites W1518885151 @default.
- W4387639503 cites W1527520754 @default.
- W4387639503 cites W1963639457 @default.
- W4387639503 cites W2022820481 @default.
- W4387639503 cites W2025596398 @default.
- W4387639503 cites W2060913109 @default.
- W4387639503 cites W2069129925 @default.
- W4387639503 cites W2080531309 @default.
- W4387639503 cites W2080792322 @default.
- W4387639503 cites W2095058792 @default.
- W4387639503 cites W2101936356 @default.
- W4387639503 cites W2128084896 @default.
- W4387639503 cites W2130275019 @default.
- W4387639503 cites W2151781750 @default.
- W4387639503 cites W2337082154 @default.
- W4387639503 cites W2419175238 @default.
- W4387639503 cites W2531147647 @default.
- W4387639503 cites W2552194003 @default.
- W4387639503 cites W2893674448 @default.
- W4387639503 cites W2919115771 @default.
- W4387639503 cites W2923537029 @default.
- W4387639503 cites W2935584311 @default.
- W4387639503 cites W2972246420 @default.
- W4387639503 cites W3006106210 @default.
- W4387639503 cites W3009480206 @default.
- W4387639503 cites W3026488073 @default.
- W4387639503 cites W3029632440 @default.
- W4387639503 cites W3035785882 @default.
- W4387639503 cites W3039811809 @default.
- W4387639503 cites W3102678975 @default.
- W4387639503 cites W3177828909 @default.
- W4387639503 cites W3204153538 @default.
- W4387639503 cites W3209165401 @default.
- W4387639503 cites W3211221363 @default.
- W4387639503 cites W3214112504 @default.
- W4387639503 cites W4206502013 @default.
- W4387639503 cites W4240189201 @default.
- W4387639503 cites W4253460544 @default.
- W4387639503 cites W4281724430 @default.
- W4387639503 cites W4288049312 @default.
- W4387639503 cites W4309334453 @default.
- W4387639503 cites W4321605930 @default.
- W4387639503 doi "https://doi.org/10.1038/s42005-023-01416-5" @default.
- W4387639503 hasPublicationYear "2023" @default.
- W4387639503 type Work @default.
- W4387639503 citedByCount "0" @default.
- W4387639503 crossrefType "journal-article" @default.
- W4387639503 hasAuthorship W4387639503A5046876703 @default.
- W4387639503 hasBestOaLocation W43876395031 @default.
- W4387639503 hasConcept C105795698 @default.
- W4387639503 hasConcept C106251023 @default.
- W4387639503 hasConcept C11413529 @default.
- W4387639503 hasConcept C121332964 @default.
- W4387639503 hasConcept C121864883 @default.
- W4387639503 hasConcept C13355873 @default.
- W4387639503 hasConcept C154945302 @default.
- W4387639503 hasConcept C159877910 @default.
- W4387639503 hasConcept C184898388 @default.
- W4387639503 hasConcept C192576344 @default.
- W4387639503 hasConcept C193999330 @default.
- W4387639503 hasConcept C199354608 @default.
- W4387639503 hasConcept C199360897 @default.
- W4387639503 hasConcept C2524010 @default.
- W4387639503 hasConcept C2777904410 @default.
- W4387639503 hasConcept C33923547 @default.
- W4387639503 hasConcept C35304006 @default.
- W4387639503 hasConcept C41008148 @default.
- W4387639503 hasConcept C48372109 @default.
- W4387639503 hasConcept C50644808 @default.
- W4387639503 hasConcept C62520636 @default.
- W4387639503 hasConcept C80444323 @default.
- W4387639503 hasConcept C94375191 @default.
- W4387639503 hasConceptScore W4387639503C105795698 @default.
- W4387639503 hasConceptScore W4387639503C106251023 @default.
- W4387639503 hasConceptScore W4387639503C11413529 @default.
- W4387639503 hasConceptScore W4387639503C121332964 @default.
- W4387639503 hasConceptScore W4387639503C121864883 @default.
- W4387639503 hasConceptScore W4387639503C13355873 @default.
- W4387639503 hasConceptScore W4387639503C154945302 @default.
- W4387639503 hasConceptScore W4387639503C159877910 @default.
- W4387639503 hasConceptScore W4387639503C184898388 @default.
- W4387639503 hasConceptScore W4387639503C192576344 @default.
- W4387639503 hasConceptScore W4387639503C193999330 @default.
- W4387639503 hasConceptScore W4387639503C199354608 @default.
- W4387639503 hasConceptScore W4387639503C199360897 @default.
- W4387639503 hasConceptScore W4387639503C2524010 @default.
- W4387639503 hasConceptScore W4387639503C2777904410 @default.
- W4387639503 hasConceptScore W4387639503C33923547 @default.
- W4387639503 hasConceptScore W4387639503C35304006 @default.
- W4387639503 hasConceptScore W4387639503C41008148 @default.
- W4387639503 hasConceptScore W4387639503C48372109 @default.
- W4387639503 hasConceptScore W4387639503C50644808 @default.
- W4387639503 hasConceptScore W4387639503C62520636 @default.
- W4387639503 hasConceptScore W4387639503C80444323 @default.